Many computer programming languages have a vast trove of reusable software components, many of which may be open source. These components can range in quality from very poor to excellent, with an equal range of performance characteristics. In many languages, there may be hundreds of thousands or even millions of different components. This poses a difficult issue for a developer: how does one select a component from a vast library?